Early brain overgrowth is a recently identified, but not yet directly measured, phenomenon in autism. Direct MRI measurement of the young autistic brain comes from five studies that found brain overgrowth at mean study ages of 2.7 to 3.9 yrs. Affected brain regions mediate higher-order social, emotional, language and cognitive functions that characterize autism. MRI studies of early brain growth in autism are absent. Indirect evidence of brain growth in autism comes from retrospective analyses of head circumference (HC). In the first study to examine the question, we reported that HC was normal average to slightly smaller than normal at birth in those who later manifested autism, but by about 12 months HC was abnormally large. Inferring brain growth rates from age-related changes in HC and placing that with the five MRI brain size studies of 2 to 4 year olds led us to the hypothesis that autism may involve a brief and agedelimited period of abnormal brain overgrowth during the first two years of life. According to the only two existent prospective studies of autism, 12 months is also approximately the first age at which autistic behavioral abnormalities first become detectable. The formation of neural circuitry is at its most exuberant and vulnerable stage during this period of development. Aberrant connectivity and neural dysfunction resulting from disruptions to this process may be key to the development of autistic behaviors. Thus, the first years of life in autism offer a unique chance to track the simultaneous emerging expression of the autistic anatomical and clinical phenotype and establish their relationship to each other and to underlying causal mechanisms, such as genes and genetic pathways that affect brain growth. We will identify early brain growth biomarkers in ASD by longitudinally MRI scanning infants at-risk for ASD, infants at-risk for developmental delay and typical infants at 12 and 24 months. We will obtain region-specific measures of volume, area, cortical thickness, fractional anisotropy and apparent diffusivity coefficient and developmental change values. Detailed anatomic maps of each region will be derived for each infant at each age. At age 36 months, a final best estimate diagnostic evaluation will identify which atrisk infants are ASD, DD or other. MRI results from these ASD and DD infants and typical infants will be statistically compared and early ASD brain growth abnormalities identified. Anatomical measures will be identified that characterize and predict the early developmental clinical phenotype of ASD that Cores B and C in collaboration with the Integrated Biostatistics Core D will identify. We will work with Project 4 and the Integrated Biostatistics Core D to use the brain growth biomarkers to discover overgrowth susceptibility genes and pathways. With Core D, we will identify separate clusters of distinctly different brain maldevelopment phenotypes among the ASD infants.